In a digital world first, Australia is barring children under 16 from social media, forcing tech giants like Meta to test an AI gatekeeper to enforce the new law. This unfolding experiment down under is sending ripples across the globe, particularly to Kenya, where regulators are keenly watching as they forge their own path on child online safety.

The heart of the matter is a new Australian law, effective December 10, that requires platforms like Instagram and Facebook to take “reasonable steps” to prevent under-16s from using their services. In response, Meta has rolled out an age-verification system. If a user tries to change their birthdate from under 16 to over 18, they must prove their age by either uploading an ID or taking a “video selfie” to be analyzed by AI.

How the AI Gatekeeper Works

Meta has partnered with Yoti, a company specializing in digital identity and age estimation, to power its video selfie analysis. The technology scans facial features to estimate a user's age. While tests show it can easily identify a clear adult, its accuracy is less certain for those near the 16-year-old threshold. Yoti itself notes that while its technology is highly accurate for teenagers, it is not perfect.

The system is designed to be privacy-preserving, with Meta and Yoti stating that the facial scan data is deleted immediately after the age check is complete. However, the rollout has not been without hitches, and even Australian officials acknowledge there will be “teething issues” with the new system. An Australian parliamentary committee has even recommended delaying the law's implementation until June 2026, citing deep reservations about data privacy.

A Precedent for Kenya?

While Australia's law is among the world's strictest, the conversation it forces is one already happening in Nairobi. The Communications Authority of Kenya (CA) recently issued its own comprehensive guidelines for child online protection, which took effect in October 2025. These rules mandate that all ICT providers must “deploy age-verification mechanisms” and incorporate safety measures into their products by design.

Unlike Australia's outright ban, Kenya's approach, as outlined by the CA, aims for “empowerment over policing,” seeking to balance a child's right to access information with robust protection. However, the guidelines explicitly require service providers to implement effective age checks, making the Australian experiment a critical case study. As one CA insider noted, while user-entered ages might be accepted initially, the ultimate goal is verification against an official ID.

• **Shared Responsibility:** The CA framework emphasizes that child safety is a shared duty among tech companies, parents, and government.
• **Safety by Design:** Companies must build safety features into products from the start, not as an afterthought.
• **Data Protection:** The new rules build on Kenya's Data Protection Act of 2019, creating a complex regulatory landscape for tech firms to navigate.

The Australian test of Meta's AI bouncer will be scrutinized by Kenyan regulators. Its successes and failures will inevitably inform how Kenya implements its own rules, shaping the digital playground for millions of young citizens.